Classification of Thyroid Follicular Lesions Based on Nuclear Texture Features
Author Information
Author(s): Metze Konradin, Ferreira Rita C, Adam Randall L
Primary Institution: University of Campinas
Hypothesis
The considerable size differences between follicular adenomas and carcinomas might have contributed to the extraordinary classification success claimed by the authors.
Conclusion
The study suggests that the classification accuracy of thyroid lesions may be influenced by the size of the lesions.
Supporting Evidence
- The study found that larger adenomas had significantly different nuclear texture features compared to carcinomas.
- With increasing adenoma size, the ability to distinguish between adenomas and carcinomas improved.
- Approximately 80% of texture parameters showed significant differences when adenomas were larger than 4.8 cm.
Takeaway
Bigger thyroid tumors look different under a microscope, which helps doctors tell them apart from smaller ones.
Methodology
Digitalized images of nuclei from histologic sections were analyzed for texture features to compare adenomas and carcinomas.
Potential Biases
The automated system was trained on a limited number of images, which may introduce bias in classification.
Limitations
The study's findings may not be applicable to smaller adenomas since they were not included in the original study.
Participant Demographics
The study involved 18 follicular adenomas and 24 microinvasive follicular carcinomas.
Statistical Information
P-Value
p<0.0001
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website